Twin–twin transfusion syndrome is a severe complication that affects about 9% of monochorionic twin pregnancies.1 Twin–twin transfusion syndrome is thought to result from an unbalanced intertwin blood flow from the donor to the recipient twin through placental vascular anastomoses. Untreated, twin–twin transfusion syndrome is associated with a high perinatal mortality and morbidity.2 Fetoscopic laser occlusion of the vascular anastomoses is currently the best treatment option for twin–twin transfusion syndrome.3 Even though laser treatment improves disease-free survival rates, adverse long-term neurodevelopment outcome is still relatively common and is a major concern for parents and physicians. The published incidence of major neurological abnormalities in twin–twin transfusion syndrome survivors after laser surgery ranges between 6% and 17%.4–7
Risk factors associated with neurodevelopment impairment have not yet been established. Most studies suggest that donor and recipient twins in twin–twin transfusion syndrome treated with laser surgery are equally at risk for cerebral injury and long-term neurodevelopment impairment.5–8 In a recent study, we found a trend toward an independent association between adverse neurodevelopment outcome and higher Quintero stage and lower gestational age at birth.7 Lack of statistical significance was probably due to the small sample size that is inherent to the rare occurrence of this twin phenomenon. To reach a larger sample size, we set up a multicenter collaboration among three European laser centers that performed long-term follow-up in twin–twin transfusion syndrome survivors after laser surgery.
The main objective of our study was to estimate the risk factors associated with adverse long-term neurodevelopment outcome in a large group of twin–twin transfusion syndrome survivors treated with fetoscopic laser surgery.
MATERIALS AND METHODS
All survivors of consecutive twin pregnancies with twin–twin transfusion syndrome treated with fetoscopic laser surgery at the Leiden University Medical Centre (the Netherlands) between August 2000 and December 2005, at the University Hospitals Leuven (Belgium) between October 2003 and August 2005 (only Belgian or Dutch residents), and at the University Hospital in Barcelona (Spain) between January 2001 and December 2003 were included in this case-control study. Part of the data from the Leiden cohort was previously included in a study on long-term outcome after laser surgery.7 These three European centers are academic referral centers specialized in intrauterine laser treatment in twin–twin transfusion syndrome pregnancies. Twin–twin transfusion syndrome was diagnosed using standard prenatal ultrasound criteria9 and staged according to the criteria of Quintero.10 The following data were recorded: gestational age at the time of laser treatment, stage of twin–twin transfusion syndrome, occurrence of twin anemia-polycythemia sequence after laser surgery, recurrence of twin–twin transfusion syndrome after laser, intrauterine fetal demise (of one or both fetuses), miscarriage, gestational age at delivery, birth weight, small for gestational age (SGA), neonatal death and infant death. Small for gestational age was defined as a birth weight below the 10th percentile for twins.11 The follow-up assessment was performed at 2 years of age (corrected for prematurity) and included a physical and neurological examination (by pediatricians) and an assessment of cognitive and neuromotor development using the Bayley Scales of Infant Development, 2nd edition (by certified and experienced psychologists). Bayley scale scores provide mental developmental indexes and psychomotor development indexes. The mean score for both mental developmental indexes and psychomotor development indexes is 100. A score below 70 is more than 2 standard deviations below the mean score and indicates a severe delay.12 Infants with very low mental developmental indexes or psychomotor development indexes scores (less than 50) were assigned a score of 49 in the database. Cerebral palsy was defined according to the European Cerebral Palsy Network and classified as diplegia, hemiplegia, quadriplegia, dyskinetic, or mixed.13 The examiners performing the cognitive and neuromotor development assessments were blinded to the natural history, but the pediatricians performing the physical and neurological evaluation were not blinded.
A composite outcome, termed neurodevelopment impairment, was defined as any of the following: cerebral palsy, mental developmental indexes score below 70, psychomotor development indexes score below 70, bilateral blindness, or bilateral deafness requiring amplification.
The institutional review board of each center approved the study, and all parents gave written, informed consent for their children.
In previous long-term follow-up study in 115 infants, we found a trend toward an independent association between lower gestational age at birth and neurodevelopment impairment (odds ratio [OR] 1.6 for each week, 95% confidence interval [CI] 0.8–3.0, P=.08).7 For this study, we calculated that a sample size of at least 270 infants was required to achieve statistical significance (with a power of 0.80, two-tailed analysis, alpha 0.05). To account for the fact that observations within twin pairs are not independent, 95% CIs for percentages were calculated using robust standard errors. Logistic regression with a “random twin effect” was applied to adjust for possible correlated effects within twins. The following potential predictors for neurodevelopment impairment were studied in a univariable logistic regression model: gestational age at laser surgery, gestational age at birth, birth weight, SGA, Quintero stage, donor (compared with recipient), intrauterine fetal demise of one twin, twin anemia-polycythemia sequence, and recurrent twin–twin transfusion syndrome. Predictors for neurodevelopment impairment that were significant (P<.05) in the univariable analysis were included in a multivariable logistic regression model to measure the independent effects. The results of the logistic models were expressed as an OR and 95% CI. P<0.05 was considered to indicate statistical significance. Analysis was performed using SPSS 12 (SPSS Inc., Chicago, IL), and the GEE module of SPSS 16 to calculate robust 95% CIs. Random effect logistic regression analysis was performed with EGRET 2.0.1 for Windows (Cytel Software Corporation, Cambridge, MA).
A total of 212 twin–twin transfusion syndrome pregnancies were treated with fetoscopic laser surgery at the three centers during the different study periods (Leiden University Medical Centre, n=113; University Hospitals Leuven, n=62; University Hospital Barcelona, n=37). Intrauterine fetal demise or miscarriage occurred in 102 of 424 fetuses (24%, 95% CI 20–29%). Neonatal death occurred in 25 of the 322 liveborn neonates (8%, 95% CI 5–12%). The incidence of twin anemia-polycythemia sequence and recurrent twin–twin transfusion syndrome after laser surgery was 4% (8of 212; 95% CI 2–7%) and 4% (9 of 212; 95% CI 2–8%), respectively. In the 16 twins with twin anemia-polycythemia sequence after laser, 2 (13%) fetuses died in utero and 2 (13%) neonates died during the neonatal period. Of the 18 fetuses with recurrence of twin–twin transfusion syndrome after laser, 6 (33%) fetuses died in utero and 2 (11%) neonates died in the neonatal period. Baseline characteristics of the studied cohort are presented in Table 1.
Two infants died after the neonatal period, and 6% of children (17 of 295) were lost to follow-up. Neurodevelopment outcome was thus assessed in 94% (278 of 295) of surviving infants.
Cerebral palsy was diagnosed in 6% of infants (17 of 278, 95% CI 4–10%) and was classified as quadriplegia (n=6), diplegia (n=3), and hemiplegia (n=8). Nineteen (7%, 95% CI 4–12%) infants had severe mental developmental delay, and 34 infants (12%, 95% CI 9–18%) had severe psychomotor developmental delay. Two infants (1%) had bilateral blindness, and two others (1%) had bilateral deafness. Overall, the incidence of neurodevelopment impairment was 18% (50 of 278, 95% CI 13–24%). The incidence of neurodevelopment impairment for infants treated in Leiden, Leuven, and Barcelona was 17% (27 of 153), 16% (12 of 77), and 23% (11 of 48). The center-to-center variation was statistically significant (P=.51, df 2). Details on the combinations of abnormal findings detected in the infants with adverse outcome are presented in Table 2.
Univariable analysis of potential risk factors for neurodevelopment impairment was performed (Table 3). Several risk factors were found to be associated with neurodevelopment impairment, including high gestational age at the time of laser surgery (OR 1.30 per week increment), high Quintero stage (OR 3.55 for each increment in stage), low gestational age at birth (OR 1.39 for each week less), and low birth weight (OR 1.18 for each 100-g decrease). We found a (small) positive correlation between gestational age at laser and Quintero stage (Pearson correlation 0.11, P=.02).
We found no statistically significant difference in neurodevelopment impairment between donors and recipients, between infants with and without SGA, and between infants with and without twin anemia-polycythemia sequence or with and without recurrent twin–twin transfusion syndrome. All surviving infants after twin anemia-polycythemia sequence or recurrent twin–twin transfusion syndrome had a normal neurodevelopment outcome.
Risk factors associated with neurodevelopment impairment by univariable analysis were entered in a multivariable logistic regression model to measure the independent association with neurodevelopment impairment (Table 3). As birth weight and gestational age at birth are closely related, birth weight was not included in the model. We found that gestational age at birth was still significantly associated with neurodevelopment impairment (OR 1.33 for each week, 95% CI 1.05–1.67, P=.016). There was a trend toward an independent association between higher Quintero stages and neurodevelopment impairment (OR 2.93 for each stage, 95% CI 0.93–9.22, P=.066).
This study shows that a low gestational age at birth is an independent risk factor for neurodevelopment impairment in twin–twin transfusion syndrome survivors. In a previous study, we were only able to detect a trend toward an independent association between low gestational age at birth and neurodevelopment impairment, probably because of a smaller sample size.7 To study a larger cohort, we opted for a multicenter study design, involving three major European fetal laser surgery centers. The main prerequisite for this study was that long-term follow-up design had to be similar in all centers, including the same standardized developmental test (the Bayley Scales of Infant Development, 2nd edition). This multicenter cooperation enabled us to evaluate the largest cohort so far of twin–twin transfusion syndrome survivors treated with laser surgery. Importantly, we were able to keep the lost-to-follow-up rate down to a minimum (6%). The association between low gestational age and neurodevelopment impairment is not surprising as prematurity is a well-recognized risk factor for adverse neurodevelopment outcome.14
In agreement with previous studies, we also found a trend toward an independent association between high Quintero stage and neurodevelopment impairment.7,15 Although the prognostic value of Quintero stages is subject of debate,16,17 our results suggest an important prognostic value of Quintero staging for neurodevelopment impairment. Increasing disease severity (ie, higher Quintero stages) may not only lead to increased perinatal mortality18 but also to increased long-term morbidity.
Interestingly, in this study, we found that, in a univariable analysis, advanced gestational age at laser is also a potential risk factor for neurodevelopment impairment. Interpretation of this association is merely speculative. As such, the fetal brain may be more vulnerable for cerebral injury (and therefore adverse neurodevelopment outcome) if laser photocoagulation is performed later on in pregnancy. Alternatively, the association between advanced gestational age at laser and neurodevelopment impairment could also partly be explained by the small but positive correlation between gestational age at laser and Quintero stage.
We found no difference in neurodevelopment impairment between donor and recipient twins, suggesting that both are at equal risk for adverse neurodevelopment outcome. These results are in agreement with previous studies.5–7,15
The rate of neurodevelopment impairment in twin–twin transfusion syndrome survivors reported in this study is high (18%), which is in agreement with our previous smaller study (17%).7 One of the limitations of this study is the absence of a control group. Adegbite et al19 studied the long-term neurodevelopment outcome in a retrospective cohort of monochorionic and dichorionic twins born between 24 and 34 weeks of gestation. They found an incidence of developmental delay, defined as a Griffith's score for mental development less than 2 standard deviations of the mean, of 15% (19 of 126) in monochorionic twin infants and 3% (5 of 142) in dichorionic twin infants. The incidence of cerebral palsy in monochorionic and dichorionic twin infants was 8% (10 of 126) and 1% (1 of 142), respectively. In a long-term follow-up study for premature infants performed recently at the Leiden University Medical Center, the rate of neurodevelopment impairment (same definition used as in this study) in the group of infants born between 27 and 32 weeks of gestation (mean gestational age of 30 weeks and mean birth weight 1,335 g) was also 16% (26 of 167).20 Thus, despite being almost 1 month older and more than 500 g heavier at birth, twin–twin transfusion syndrome survivors in this cohort have similar rate of neurodevelopment impairment as very preterm infants. Obstetricians and pediatricians should thus recognize the considerable risks for the poor developmental outcomes for twin–twin transfusion syndrome survivors.
Another potential limitation of this study is related to the predictive value of the Bayley test. Although the Bayley test is a valid discriminative measure recommended for identification of motor and mental delay or eligibility for early intervention service, its predictive value for future motor and mental outcome is limited.
The rate of neurodevelopment impairment in twin–twin transfusion syndrome stage I found in this study is very low (3%). In the randomized trial published in 2004, Senat et al3 showed that disease-free survival rate was higher after laser treatment in all stages (including stages I and II). Nevertheless, a plea has recently been made for another randomized controlled trial to determine the best treatment option in twin–twin transfusion syndrome with low stages (I and/or II).23,24 Whether the evidence, so far, is sufficient to warrant such a trial is currently being debated.25 Regardless of this controversy, if such a trial will be conducted in the near future, primary outcome should be long-term disease-free survival. Our study provides important data that can be used in the power analysis to determine the sample size needed in a “stage I trial.”
In conclusion, we were able to establish several risk factors for neurodevelopment impairment in twin–twin transfusion syndrome survivors after laser surgery, including higher gestational age at laser, higher Quintero stage, lower gestational at birth, and low birth weight. Although thorough follow-up is recommended in all twin–twin transfusion syndrome-survivors, particular attention is required for the high-risk groups as defined in this study.
1. Lewi L, Jani J, Blickstein I, Huber A, Gucciardo L, Van Mieghem T, et al. The outcome of monochorionic diamniotic twin gestations in the era of invasive fetal therapy: a prospective cohort study. Am J Obstet Gynecol 2008;199:514.e1–8.
2. van Gemert MJ, Umur A, Tijssen JG, Ross MG. Twin-twin transfusion syndrome: etiology, severity and rational management. Curr Opin Obstet Gynecol 2001;13:193–206.
3. Senat MV, Deprest J, Boulvain M, Paupe A, Winer N, Ville Y. Endoscopic laser surgery versus serial amnioreduction for severe twin-to-twin transfusion syndrome. N Engl J Med 2004;351:136–44.
4. Sutcliffe AG, Sebire NJ, Pigott AJ, Taylor B, Edwards PR, Nicolaides KH. Outcome for children born after in utero laser ablation therapy for severe twin-to-twin transfusion syndrome. BJOG 2001;108:1246–50.
5. Banek CS, Hecher K, Hackeloer BJ, Bartmann P. Long-term neurodevelopmental outcome after intrauterine laser treatment for severe twin-twin transfusion syndrome. Am J Obstet Gynecol 2003;188:876–80.
6. Graef C, Ellenrieder B, Hecher K, Hackeloer BJ, Huber A, Bartmann P. Long-term neurodevelopmental outcome of 167 children after intrauterine laser treatment for severe twin-twin transfusion syndrome. Am J Obstet Gynecol 2006;194:303–8.
7. Lopriore E, Middeldorp JM, Sueters M, Oepkes D, Vandenbussche FP, Walther FJ. Long-term neurodevelopmental outcome in twin-to-twin transfusion syndrome treated with fetoscopic laser surgery. Am J Obstet Gynecol 2007;196:231–4.
8. Lopriore E, Wezel-Meijler G, Middeldorp JM, Sueters M, Vandenbussche FP, Walther FJ. Incidence, origin, and character of cerebral injury in twin-to-twin transfusion syndrome treated with fetoscopic laser surgery. Am J Obstet Gynecol 2006;194:1215–20.
9. Huber A, Hecher K. How can we diagnose and manage twin-twin transfusion syndrome? Best Pract Res Clin Obstet Gynaecol 2004;18:543–56.
10. Quintero RA, Morales WJ, Allen MH, Bornick PW, Johnson PK, Kruger M. Staging of twin-twin transfusion syndrome. J Perinatol 1999;19:550–5.
11. Blickstein I. Is it normal for multiples to be smaller than singletons? Best Pract Res Clin Obstet Gynaecol 2004;18:613–23.
12. Hack M, Taylor HG, Drotar D, Schluchter M, Cartar L, Wilson-Costello D, et al. Poor predictive validity of the Bayley Scales of Infant Development for cognitive function of extremely low birth weight children at school age. Pediatrics 2005;116:333–41.
13. Surveillance of cerebral palsy in Europe: a collaboration of cerebral palsy surveys and registers. Surveillance of Cerebral Palsy in Europe (SCPE). Dev Med Child Neurol 2000;42:816–24.
14. Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet 2008;371:261–9.
15. Lopriore E, Nagel HT, Vandenbussche FP, Walther FJ. Long-term neurodevelopmental outcome in twin-to-twin transfusion syndrome. Am J Obstet Gynecol 2003;189:1314–9.
16. Taylor MJ, Govender L, Jolly M, Wee L, Fisk NM. Validation of the Quintero staging system for twin-twin transfusion syndrome. Obstet Gynecol 2002;100:1257–65.
17. Jain V, Fisk NM. The twin-twin transfusion syndrome. Clin Obstet Gynecol 2004;47:181–202.
18. Middeldorp JM, Sueters M, Lopriore E, Klumper FJ, Oepkes D, Devlieger R, et al. Fetoscopic laser surgery in 100 pregnancies with severe twin-to-twin transfusion syndrome in the Netherlands. Fetal Diagn Ther 2007;22:190–4.
19. Adegbite AL, Castille S, Ward S, Bajoria R. Neuromorbidity in preterm twins in relation to chorionicity and discordant birth weight. Am J Obstet Gynecol 2004;190:156–63.
20. Rijken M, Stoelhorst GM, Martens SE, van Zwieten PH, Brand R, Wit JM, et al. Mortality and neurologic, mental, and psychomotor development at 2 years in infants born less than 27 weeks' gestation: the Leiden follow-up project on prematurity. Pediatrics 2003;112:351–8.
21. Tieman BL, Palisano RJ, Sutlive AC. Assessment of motor development and function in preschool children. Ment Retard Dev Disabil Res Rev 2005;11:189–96.
22. Eight-year outcome in infants with birth weight of 500 to 999 grams: continuing regional study of 1979 and 1980 births. Victorian Infant Collaborative Study Group. J Pediatr 1991;118:761–7.
23. O'Donoghue K, Cartwright E, Galea P, Fisk NM. Stage I twin-twin transfusion syndrome: rates of progression and regression in relation to outcome. Ultrasound Obstet Gynecol 2007;30:958–64.
24. Roberts D, Neilson JP, Kilby M, Gates S. Interventions for the treatment of twin-twin transfusion syndrome. Cochrane Database Syst Rev 2008;CD002073.
25. Ville Y. Twin-to-twin transfusion syndrome: time to forget the Quintero staging system? Ultrasound Obstet Gynecol 2007;30:924–7.